Dark energy is a phrase used by physicists to describe a mysterious'something' that is causing the universe to accelerate in its expansion. It is the'gravitational glue' that holds galaxies together and is thought to make up five sixths of the universe's mass. These substances have profound effects on the birth and lives of galaxies and stars and yet almost nothing is known about their physical nature. But now a new computer model, twenty-five times faster than other methods, will allow scientists to compute virtual universes in the search of explanations about these mysteries. The new method makes the universe models more accurate by comparing the model's properties with an'inverted' version.

Linguists and primatologists have joined forces to create the groundwork for'primate linguistics,' helping to decipher the meanings behind monkey speech. The comprehensive study examines the calls of different species, analyzing the structure and placement of these vocalizations, and explains what individual calls and sequences mean. While the language of primates may not be as complex as our own, researchers say these animals demonstrate linguistic capabilities that are both'exciting and sometimes challenging.' Linguists and primatologists have joined forces to create the groundwork for'primate linguistics,' helping to decipher the meanings behind monkey speech. The study, led by an international team of researchers, was published recently in the journals Natural Language & linguistic Theory, and builds on earlier research.

Deep clustering is a recently introduced deep learning architecture that uses discriminatively trained embeddings as the basis for clustering. It was recently applied to spectrogram segmentation, resulting in impressive results on speaker-independent multi-speaker separation. In this paper we extend the baseline system with an end-to-end signal approximation objective that greatly improves performance on a challenging speech separation. We first significantly improve upon the baseline system performance by incorporating better regularization, larger temporal context, and a deeper architecture, culminating in an overall improvement in signal to distortion ratio (SDR) of 10.3 dB compared to the baseline of 6.0 dB for two-speaker separation, as well as a 7.1 dB SDR improvement for three-speaker separation. We then extend the model to incorporate an enhancement layer to refine the signal estimates, and perform end-to-end training through both the clustering and enhancement stages to maximize signal fidelity. We evaluate the results using automatic speech recognition. The new signal approximation objective, combined with end-to-end training, produces unprecedented performance, reducing the word error rate (WER) from 89.1% down to 30.8%. This represents a major advancement towards solving the cocktail party problem.

We present a general framework for classifying partially observed dynamical systems based on the idea of learning in the model space. In contrast to the existing approaches using model point estimates to represent individual data items, we employ posterior distributions over models, thus taking into account in a principled manner the uncertainty due to both the generative (observational and/or dynamic noise) and observation (sampling in time) processes. We evaluate the framework on two testbeds - a biological pathway model and a stochastic double-well system. Crucially, we show that the classifier performance is not impaired when the model class used for inferring posterior distributions is much more simple than the observation-generating model class, provided the reduced complexity inferential model class captures the essential characteristics needed for the given classification task.

We investigate a long-debated question, which is how to create predictive models of recidivism that are sufficiently accurate, transparent, and interpretable to use for decision-making. This question is complicated as these models are used to support different decisions, from sentencing, to determining release on probation, to allocating preventative social services. Each use case might have an objective other than classification accuracy, such as a desired true positive rate (TPR) or false positive rate (FPR). Each (TPR, FPR) pair is a point on the receiver operator characteristic (ROC) curve. We use popular machine learning methods to create models along the full ROC curve on a wide range of recidivism prediction problems. We show that many methods (SVM, Ridge Regression) produce equally accurate models along the full ROC curve. However, methods that designed for interpretability (CART, C5.0) cannot be tuned to produce models that are accurate and/or interpretable. To handle this shortcoming, we use a new method known as SLIM (Supersparse Linear Integer Models) to produce accurate, transparent, and interpretable models along the full ROC curve. These models can be used for decision-making for many different use cases, since they are just as accurate as the most powerful black-box machine learning models, but completely transparent, and highly interpretable.

Remember those little six-wheeled robots we told you about in April? They're now set for a commercial rollout in London and three other European cities. The robots, from Starship Technologies, will be deployed this month to make deliveries for food-ordering services Just Eat and Pronto, and carry packages for courier service Hermes and supermarket Metro Group. The Starship delivery bots will be stationed at kitchens, delivery hubs, and supermarkets in London, Düsseldorf, Bern, and Hamburg. When an order comes in, the bots will drive themselves to collect their cargo, store it in their holds (which can take about two shopping bags' worth of stuff), then trundle on to their destinations.

Self-driving robots will soon start delivering food and groceries. This month Starship Technologies is rolling out its six-wheeled delivery robots in London, Dusseldorf, Bern, and Hamburg, Quartz reports. The robots will be used by two food delivery services in those areas, Just Eat and Pronto, as well as courier service Hermes and grocery store Metro Group. Starship hopes that this new technology will help cut both the time and costs associated with delivery. The self-driving robots won't exactly be self-driving at first.

Moodstocks, a Parisian startup that develops image recognition tools for smartphones, is joining Google. The companies announced the acquisition today, sans financials. Around since 2008, Moodstocks hasn't had considerable traction. But the company has tech and engineers working on machine learning, something Google cannot get enough of as it competes with rivals like Apple and Facebook for talent. And Moodstocks' core service -- "to give eyes to machines by turning cameras into smart sensors," as its parting note described -- fits with Google's vision for image search and augmented reality, where a phone (or something else) knows your physical surroundings. Also, the acquisition price may have been low thanks to wobbling global markets.

On 7 May, a Tesla Model S was involved in a fatal accident in Florida. At the time of the accident, the vehicle was driving itself, using its Autopilot system. The system didn't stop for a tractor-trailer attempting to turn across a divided highway, and the Tesla collided with the trailer. In a statement, Tesla Motors said this is the "first known fatality in just over 130 million miles [210 million km] where Autopilot was activated" and suggested that this ratio makes the Autopilot safer than an average vehicle. Early this year, Tesla CEO Elon Musk told reporters that the Autopilot system in the Model S was "probably better than a person right now."

An eye hospital in London is entering the debate over patient privacy by sharing images of patients' retinas with a Google-owned artificial intelligence project. Moorfields Eye Hospital is anonymizing and then sharing patient information with DeepMind, a machine-learning AI company that plans to use the hospital's non-invasive retina scans to train its machines, which must scan thousands and thousands of images to "learn" how an eye should look. While patients consented to general research, privacy advocates have expressed concerns they may not have realized the extent to which their personal information – even scans of their eyes – would be handed over to an outside party such as Google. The hospital has taken an important first step by informing patients before the project begins, which is likely a lesson learned from a previous medical research project in Britain. In a previous data-sharing partnership between DeepMind and three other London hospitals, patients discovered the involvement of their data only haphazardly afterward, the BBC reported.